grilling

Interviews users in structured question rounds to stress-test plans and decisions.

Updated May 31, 2026
One-click install
npx skills add https://github.com/AlexanderNarbaev/agi --skill grilling-alexandernarbaev
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: grilling
Source: https://github.com/AlexanderNarbaev/agi/tree/main/.opencode/skills/matt-pocock/grilling
Command: npx skills add https://github.com/AlexanderNarbaev/agi --skill grilling-alexandernarbaev

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Plans and decisions often contain hidden assumptions and unresolved dependencies that only surface after execution begins. This Skill systematically interrogates a plan, decision, or idea by mapping it as a design tree and asking every reachable question until nothing is left silently assumed. ## Core Features & Use Cases - Design Tree Mapping: Models every decision as a branch with dependent sub-decisions, so the full scope of open questions is visible. - Frontier-Based Rounds: Asks only the questions whose prerequisites are settled, numbering each one and providing a recommended answer, then recomputes the frontier after each round. - Fact-Finding via Sub-Agents: Dispatches sub-agents to look up facts from the filesystem or tools instead of asking the user for information that can be retrieved directly. - Use Case: Before starting a major refactor, have the Skill grill you on architecture choices, migration order, and rollback strategy until every dependency between decisions is resolved and confirmed. ## Quick Start Ask the assistant to grill you on your plan for an upcoming project or decision and answer each numbered round of questions until the frontier is empty.

Frequently Asked Questions about grilling

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I stress-test a plan or idea with an AI assistant?

Invoke the grilling skill and describe your plan or decision. The assistant maps it as a design tree and interviews you in rounds of numbered questions, each with a recommended answer, until every branch is resolved.

What is a design tree in decision planning?

A design tree models each decision as a node whose sub-decisions branch off it. The skill only asks questions whose prerequisite decisions are already settled, so you never answer questions that depend on unresolved choices.

How does the frontier-based questioning round work?

Each round asks all currently answerable questions at once, numbered with recommended answers. Your responses settle decisions, unblock dependent questions, and trigger the next round until the frontier is empty.

Does the skill ask the user for facts it could look up itself?

No. When a question needs a fact from the filesystem or tools, the skill dispatches a sub-agent to find it. Only questions downstream of that fact wait; the rest of the frontier is asked immediately.

When is a grilling session considered complete?

The session ends when the frontier is empty, meaning every branch of the design tree has been visited and no assumption remains unexamined. The skill waits for explicit user confirmation before acting on the shared understanding.